**Note this summary focuses on data from the Baltic for 2018 submitted in 2020. Values used are hours fished.
If anonymisation is required (i.e. if there are csquares in the aggregated dataset where there are 2 or fewer unique vessels) the following procedure is undertaken to ensure that there are allways 3 or more vessels contributing to any value in the dataset. The anonimysation procedure follows two steps:
during this process intial ‘breaks’ are defined giving an upper and lower value for group_by
Alternatives * Groups could be combined with the closed group in value to it. * The bin size of step 1 can be varied and may give more accurate results (less grouping) if a smaller bin size is used.
Example using Benthis gear groupings
Example using Fishing category
A summary of the differences between values for those that require to be anonymised.
Min. 1st Qu. Median Mean 3rd Qu. Max.
0.00 0.03 0.03 0.76 0.13 227.40
Figure 1: Number of vessels per category when the 75th percentile is used to set the bin width
This section summarises the impact on the published value due to replacing the submitted value with a summary, in this case the middle of the range.
alternatives The mean of the values could be used, rather than the middle of the group.
Figure 2: published value versus the submitted values. Note, many group will have 3 vessels (by design).
Figure 3: The diffrence between the published value and the submitted value. Grouping tends to affect larger values more.